AutoTool: Dynamic Tool Selection and Integration for Agentic Reasoning
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912765104357376 |
|---|---|
| author | Zou, Jiaru Yang, Ling Qi, Yunzhe Chen, Sirui Ai, Mengting Shen, Ke He, Jingrui Wang, Mengdi |
| author_facet | Zou, Jiaru Yang, Ling Qi, Yunzhe Chen, Sirui Ai, Mengting Shen, Ke He, Jingrui Wang, Mengdi |
| contents | Agentic reinforcement learning has advanced large language models (LLMs) to reason through long chain-of-thought trajectories while interleaving external tool use. Existing approaches assume a fixed inventory of tools, limiting LLM agents' adaptability to new or evolving toolsets. We present AutoTool, a framework that equips LLM agents with dynamic tool-selection capabilities throughout their reasoning trajectories. We first construct a 200k dataset with explicit tool-selection rationales across 1,000+ tools and 100+ tasks spanning mathematics, science, code generation, and multimodal reasoning. Building on this data foundation, AutoTool employs a dual-phase optimization pipeline: (i) supervised and RL-based trajectory stabilization for coherent reasoning, and (ii) KL-regularized Plackett-Luce ranking to refine consistent multi-step tool selection. Across ten diverse benchmarks, we train two base models, Qwen3-8B and Qwen2.5-VL-7B, with AutoTool. With fewer parameters, AutoTool consistently outperforms advanced LLM agents and tool-integration methods, yielding average gains of 6.4% in math & science reasoning, 4.5% in search-based QA, 7.7% in code generation, and 6.9% in multimodal understanding. In addition, AutoTool exhibits stronger generalization by dynamically leveraging unseen tools from evolving toolsets during inference. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_13278 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | AutoTool: Dynamic Tool Selection and Integration for Agentic Reasoning Zou, Jiaru Yang, Ling Qi, Yunzhe Chen, Sirui Ai, Mengting Shen, Ke He, Jingrui Wang, Mengdi Computation and Language Machine Learning Agentic reinforcement learning has advanced large language models (LLMs) to reason through long chain-of-thought trajectories while interleaving external tool use. Existing approaches assume a fixed inventory of tools, limiting LLM agents' adaptability to new or evolving toolsets. We present AutoTool, a framework that equips LLM agents with dynamic tool-selection capabilities throughout their reasoning trajectories. We first construct a 200k dataset with explicit tool-selection rationales across 1,000+ tools and 100+ tasks spanning mathematics, science, code generation, and multimodal reasoning. Building on this data foundation, AutoTool employs a dual-phase optimization pipeline: (i) supervised and RL-based trajectory stabilization for coherent reasoning, and (ii) KL-regularized Plackett-Luce ranking to refine consistent multi-step tool selection. Across ten diverse benchmarks, we train two base models, Qwen3-8B and Qwen2.5-VL-7B, with AutoTool. With fewer parameters, AutoTool consistently outperforms advanced LLM agents and tool-integration methods, yielding average gains of 6.4% in math & science reasoning, 4.5% in search-based QA, 7.7% in code generation, and 6.9% in multimodal understanding. In addition, AutoTool exhibits stronger generalization by dynamically leveraging unseen tools from evolving toolsets during inference. |
| title | AutoTool: Dynamic Tool Selection and Integration for Agentic Reasoning |
| topic | Computation and Language Machine Learning |
| url | https://arxiv.org/abs/2512.13278 |